Machine learning predicts the functional composition of the protein corona and the cellular recognition of nanoparticles

Abstract

Protein corona formation is critical for the design of ideal and safe nanoparticles (NPs) for nanomedicine, biosensing, organ targeting, and other applications, but methods to quantitatively predict the formation of the protein corona, especially for functional compositions, remain unavailable. The traditional linear regression model performs poorly for the protein corona, as measured by R2 (less than 0.40). Here, the performance with R2 over 0.75 in the prediction of the protein corona was achieved by integrating a machine learning model and meta-analysis. NPs without modification and surface modification were identified as the two most important factors determining protein corona formation. According to experimental verification, the functional protein compositions (e.g., immune proteins, complement proteins, and apolipoproteins) in complex coronas were precisely predicted with good R2 (most over 0.80). Moreover, the method successfully predicted the cellular recognition (e.g., cellular uptake by macrophages and cytokine release) mediated by functional corona proteins. This workflow provides a method to accurately and quantitatively predict the functional composition of the protein corona that determines cellular recognition and nanotoxicity to guide the synthesis and applications of a wide range of NPs by overcoming limitations and uncertainty.

Complexity of the Nano-Bio Interface and the Tortuous Path of Metal Oxides in Biological Systems

“…Compared to other traditional linear regression approaches with poor prediction performance (R2 values less than 0.40), machine learning can achieve good prediction performance especially when meta-analysis is incorporated into the model. Using this approach, machine learning models predicted the content of the protein corona on a given nanoparticle with an R2 of 0.75 [ 181 ]. In addition, the model successfully predicted the biological outcomes of the corona.…”

“…While this field is still in its infancy and approaches such as MaSIF have yet to be applied to interactions at the bio-nano interface, these approaches may prove to be powerful. As researchers begin to expand and improve the data sets associated with different nanomaterials, machine learning approaches will become more accurate and useful in designing biocompatible nanomaterials with predictable biological and therapeutic actions [ 181 ].…”

Current Status and Challenges Associated with CNS-Targeted Gene Delivery across the BBB

“…Once nanoparticles are injected into the systemic circulation, nanoparticles encounter serum components, such as proteins, resulting in the formation of a protein corona on the surface. The formation of a protein corona is critical for the design of efficient and safe nanoparticles for tissue-targeting, nanomedicines, and other applications, so research related to the protein corona is a subject of great interest [ 136 , 137 , 138 , 139 , 140 , 141 , 142 , 143 , 144 , 145 , 146 ]. Although protein corona formation on a nanoparticle surface may adversely affect targeting [ 147 ], controlling them can also be applied to achieve more effective targeting [ 51 , 54 , 55 , 148 , 149 , 150 , 151 , 152 , 153 , 154 ].…”

Synthesis, characterization, and biological applications of semiconducting polythiophene‐based nanoparticles

“…The evaluation of the surface properties and structural characteristics of NPs is also of primary importance to study the composition, aggregation, and bonding nature of the constituting nanomaterial 78 . The surface chemistry of NPs plays a pivotal role in dictating their behavior both in vitro and in vivo 79 . SEM/TEM‐EDX and Zeta potential measurements, as aforementioned, can give information on the elemental composition and charge of the NPs surface, respectively; however, other techniques such as nuclear magnetic resonance (NMR) and X‐ray photoelectron spectroscopies (XPSs) are fundamental to fully elucidate the NPs’ surface composition and properties.…”